Individual differences in faking integrity tests.

Abstract

The present study assessed whether success at faking a commercially available integrity test relates to individual differences among the test takers. We administered the Reid Report, an overt integrity test, twice to a sample of college students with instructions to answer honestly on one administration and "fake good" on the other. These participants also completed a measure of general cognitive ability, the Raven Advanced Progressive Matrices. Integrity test scores were 1.3 standard deviations higher in the faking condition (p<.05). There was a weak, but significant, positive relation between general cognitive ability and faking success, calculated as the difference in scores between the honest and faked administrations of the Reid Report (r=.17, p<.05). An examination of the correlations between faking success and general cognitive ability by item type suggested that the relation is due to the items that pose hypothetical scenarios, e.g., "Should an employee be fired for stealing a few office supplies?" (r=.22, p<.05) and not the items that ask for admissions of undesirable past behaviors. e.g., "Have you ever stolen office supplies?" (r=.02, p>.05: t=2.06, p<.05) for the difference between correlations. These results suggest that general cognitive ability is indeed an individual difference relevant to success at faking an overt integrity test.